Skip to content
Deepbox
08intermediate

Logistic Regression

Binary classification with logistic regression on the Iris dataset: setosa vs the other two species.

What you will learn

  • Use deepbox/datasets for loadIris.
  • Use deepbox/ml for LogisticRegression.
  • Use deepbox/metrics for accuracy, precision, recall, f1Score, confusionMatrix.
  • Use deepbox/preprocess for trainTestSplit, StandardScaler.
  • Binary classification with logistic regression on the Iris dataset: setosa vs the other two species.

Source

index.ts
/** * Example 08: Logistic Regression * * Binary classification with logistic regression on the Iris dataset: * setosa vs the other two species. Features are standardized first, and the * model is scored with accuracy, precision, recall, F1 and a confusion matrix. */ import { loadIris } from "deepbox/datasets";import { accuracy, confusionMatrix, f1Score, precision, recall } from "deepbox/metrics";import { LogisticRegression } from "deepbox/ml";import { StandardScaler, trainTestSplit } from "deepbox/preprocess"; console.log("=== Logistic Regression ===\n"); // Load the famous Iris dataset for classificationconst iris = loadIris();console.log(`Dataset: ${iris.data.shape[0]} samples, ${iris.data.shape[1]} features\n`); // Simplify to binary classification: setosa (0) vs non-setosa (1).// Clipping the labels 0, 1, 2 to the range [0, 1] maps setosa to 0 and the others to 1.const y = iris.target.clip(0, 1); // Split into train (70%) and test (30%) setsconst [XTrain, XTest, yTrain, yTest] = trainTestSplit(iris.data, y, {  testSize: 0.3,  randomState: 42,}); console.log(`Training set: ${XTrain.shape[0]} samples`);console.log(`Test set: ${XTest.shape[0]} samples\n`); // Standardize features (mean 0, std 1) so gradient descent converges faster.// Fit the scaler on the training set only, then apply it to both sets.const scaler = new StandardScaler();scaler.fit(XTrain);const XTrainScaled = scaler.transform(XTrain);const XTestScaled = scaler.transform(XTest); console.log("Features scaled\n"); // Create and train logistic regression classifierconst model = new LogisticRegression({ maxIter: 1000, learningRate: 0.1 });model.fit(XTrainScaled, yTrain); console.log("Model trained.\n"); // Make predictions on test dataconst yPred = model.predict(XTestScaled); // Calculate classification metricsconst acc = accuracy(yTest, yPred);const prec = precision(yTest, yPred);const rec = recall(yTest, yPred);const f1 = f1Score(yTest, yPred); console.log("Model Performance:");console.log(`Accuracy:  ${(acc * 100).toFixed(2)}%`);console.log(`Precision: ${(prec * 100).toFixed(2)}%`);console.log(`Recall:    ${(rec * 100).toFixed(2)}%`);console.log(`F1-Score:  ${(f1 * 100).toFixed(2)}%\n`); const cm = confusionMatrix(yTest, yPred);console.log("Confusion Matrix:");console.log(cm.toString());

Output

Console output
Console output only: accuracy, precision, recall, F1 and the confusion matrix.